Unverified50% confidenceOpinionExact time
作者认为RAG知识库是智能体开发中投入产出比最高的方向，因为几乎每家公司都需要做知识库
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/17/2026
First Seen
Valid until: 10/15/2026
Sources
Related Claims
Unverified与RAG强调文档检索不同，Skill更强调对特定任务的过程性知识封装68% similarUnverifiedThe skill system achieves automatic discovery through RAG (Retrieval-Augmented Generation) technology by vectorizing and storing descriptions of all skills, and automatically injecting a skill into the current context when semantic similarity exceeds a threshold.67% similarUnverifiedSurfaceFree's "Expert Mode" injects structured knowledge such as industry best practices, field specifications, and process templates into the generation process through RAG (Retrieval-Augmented Generation) technology.66% similarUnverifiedSkill与RAG的区别在于Skill是显式调用的结构化知识，而RAG是隐式检索的非结构化知识65% similarUnverifiedPractitioners in the engineering implementation direction need to be familiar with skills such as prompt engineering, RAG, Agent orchestration frameworks (such as LangChain and LlamaIndex), and model deployment and serving62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/544285API
curl https://kongchang.com/api/v1/knowledge/claims/544285MCP
get_claim(id=544285)